Our long-term goals are to gain a deeper understanding of working memory, which seems to be at the core of many higher-level cognitive processes, and to determine the role played by the working memory system in age- related declines in cognition. Toward these ends, we propose to use a combination of neuroimaging (PET) and behavioral studies with younger and older adults. Our specific objectives are to study how younger and older adults differ -- both cognitively and neuroanatomically -- with respect to: (1) the verbal and spatial buffers of working memory; (2) executive processes that contextually code the contents of working memory; (3) executive processes required to manage multiple goals stored in working memory, and (4) executive processes involved when complex sequential operations are performed on the contents of working memory . For Objective (1), we will employ both spatial and verbal short-term storage tasks; e.g., determine whether the position of a current item matches that of a previously presented item, or determine whether the identity of a current item matches that of a previous one. To pursue Objective (2), we will use tasks that require subjects to remember the temporal order of items as well as their contents in working memory, e.g., determine whether the current item is identical to the one presented two back. Regarding Objective (3), we will require subjects to concurrently perform two related tasks, e.g., decide whether each of a series of statements is true and false, and remember some target words from each sentence for subsequent recall. For Objective (4), we will employ tasks in which subjects have to carry out, mentally, a complex series of numerical calculations. We expect that aging will have less deleterious effects on tasks that require only storage than on tasks that require executive processes as well. In addition, the aging decline should be more notable with more demanding executive processes. Furthermore, we expect that aging declines in storage and executive processes will be associated with specific changes in brain activation. For example, there is reason to expect that a decline in temporal coding ability might be accompanied by reduced activation in specific areas of dorsolateral prefrontal cortex; a decline in managing multiple goals might be accompanied by activation changes in the same prefrontal region or in others.